1. Field of the Invention
The invention relates to a method for effecting the calibration of a mechanical pressure gauge to achieve a required degree of linearity between movement of the pointer of the gauge and the applied pressure.
2. Summary of the Prior Art
Mechanical pressure gauges consist of a pressure measuring element and a mechanism (commonly called a "movement") which converts the generally linear motion of the pressure movable element into rotary motion of a pinion upon which is mounted a dial indicating pointer which indicates pressure on a circular dial. Because of the geometric relationship between the moving pressure element and the rotational movement of the pinion, the rotation of the pointer is not linear with respect to the motion of the pressure movable element. Typical gauges of this type are shown in U.S. Pat. Nos. 4,055,085, 4,148,123, 4,240,298, 4,361,046, and 4,542,654. For each of the typical pressure gauges shown in the aforementioned patents, the best degree of non-linearity of pointer rotation which can be achieved approximates a sine wave as shown in the graph of FIG. 1. As shown there, the error varies from zero at the zero pressure and maximum pressure ends of the pointer dial and progresses in a sine wave between these two points with another zero being recorded at the midway point or 135.degree. position for a gauge with a 270.degree. range. (Not all gauges have a 270.degree. dial). The permissible non-linearity of highly accurate industrial pressure gauges is preferably less than one quarter of one percent. If this degree of non-linearity, or less, is obtained, then the gauge will readily meet the accuracy requirements of the industry which are generally on the order of not more than one half of one percent. Unacceptable calibration curves are also shown in FIG. 1, even though the calibration error is zero at full scale, illustrating the importance of checking the accuracy of the calibration at one or more pressures intermediate zero and full scale.
To obtain the ideal "baseline" curve, a variety of adjustments have heretofore been employed, such as rotating the movement mechanism, bending the tip of the Bourdon type pressure element, or adjusting the length or span of a connecting link between the pressure element and the segment gear. Any one of these adjustments, even though limited to very. small magnitudes, may drastically effect the non-linearity of the pressure scale at mid scale by as much as one half of one percent. Any one adjustment generally requires another variable to be adjusted.
Each of the aforementioned patents mention the calibration problem. As stated in U.S. Pat. No., 4,148,123:
"It is obvious that every adjustment of the span necessarily changes linearity and that every adjustment in linearity in turn influences the span so that as a rule the span and linearity have to be adjusted several times in an alternating fashion until the desired display tolerances are attained. It is self evident that the conventional adjustment procedure is thus a time consuming and expensive operation."
Calibration for linearity was accomplished only by skilled operators who could, by experience, judge the type and amount of adjustments needed, make the adjustments, and then check the position of the pointer at zero, quarter scale, mid scale, three quarter scale and full scale to determine whether the proper adjustment had been made. Many iterations of this procedure were required in order to achieve the desired accuracy. Obviously, if the amount of movement rotation, tip bending or span adjustment were known in order to change any midpoint non-linearity to the desired zero level, the known amount of adjustment could be made without the need to readjust and recheck the linearity many times.